What Is Monte Carlo Simulation? What Is Sequence of Returns Risk?
How Monte Carlo simulation works, what sequence of returns risk is, and what these concepts look like with Indian market data.
The Gap Between Average and Actual Returns
The Nifty 50’s long-term average CAGR is ~12%. Year-by-year reality looks different:
| Year | Nifty 50 Return |
|---|---|
| 2007 | +54% |
| 2008 | -52% |
| 2009 | +76% |
| 2011 | -25% |
| 2017 | +29% |
| 2021 | +24% |
A retirement projection using a flat 12% every year produces a single line. The actual path — with crashes and recoveries — produces a wide range of possible outcomes. Monte Carlo simulation maps that range.
How Monte Carlo Simulation Works
┌─────────────────────────────────────────┐
│ INPUTS │
│ • Starting corpus: ₹3 Cr │
│ • Annual withdrawal: ₹10.5L │
│ • Time horizon: 30 years │
│ • Asset allocation: 60/40 equity/debt │
│ • Return distribution: mean, std dev │
│ from historical data │
└──────────────────┬──────────────────────┘
▼
┌─────────────────────────────────────────┐
│ SIMULATION ENGINE │
│ For each of 10,000 runs: │
│ For each year (1 to 30): │
│ 1. Draw random return from │
│ distribution │
│ 2. Apply return to portfolio │
│ 3. Subtract withdrawal │
│ 4. If portfolio ≤ 0 → failure │
└──────────────────┬──────────────────────┘
▼
┌─────────────────────────────────────────┐
│ OUTPUT │
│ • Success rate: X% of runs survived │
│ • Percentile outcomes at year 30 │
│ • Distribution of portfolio values │
└─────────────────────────────────────────┘Each run uses a different random sequence of annual returns. The returns are drawn from a distribution matching historical data (e.g., Indian equity: mean ~12%, standard deviation ~25%).
What the Output Looks Like
₹3 Cr corpus. ₹10.5L/year withdrawal (3.5%). 60/40 equity/debt. 30-year horizon.
Portfolio Value at Year 30 (₹ Cr)
10 | ···· 95th percentile
| ·····
8 | ······
| ······
6 | ······ ---- 75th percentile
| ------
4 | ------
| ----- ════ Median (50th)
2 |════════════════════════
|▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓ ▓▓▓▓ 25th percentile
0 |████████████████████████████████████████ ████ 10th percentile
└──────────────────────────────────────
0 5 10 15 20 25 30 years| Percentile | Portfolio at Year 30 | Survived? |
|---|---|---|
| 95th (best runs) | ~₹10 Cr | Yes |
| 75th | ~₹6 Cr | Yes |
| 50th (median) | ~₹3.5 Cr | Yes |
| 25th | ~₹1.2 Cr | Yes (barely) |
| 10th | ₹0 (exhausted ~year 24) | No |
| 5th | ₹0 (exhausted ~year 20) | No |
A flat-return calculator using the same 12% average shows a single outcome: ~₹8 Cr surplus. The Monte Carlo output shows that ~10–15% of paths run out of money.
What Is Sequence of Returns Risk (SORR)
SORR is the effect of the order in which returns occur. Two portfolios with identical average returns over 20 years can have very different outcomes depending on whether bad years happen early or late.
Same Average, Different Sequence
₹3 Cr starting corpus. ₹10.5L/year withdrawal. Same average return over 20 years.
Sequence A — Bad years early:
Year: 1 2 3 4 5 6-20
Return: -30% -15% +5% -10% +8% strong recoverySequence B — Bad years late:
Year: 1-15 16 17 18 19 20
Return: strong growth -10% +5% -15% -30% +8%| Sequence A (bad early) | Sequence B (bad late) | |
|---|---|---|
| Portfolio after year 5 | ~₹1.4 Cr | ~₹7.2 Cr |
| Portfolio after year 20 | ₹0 (exhausted) | ~₹3.8 Cr |
Why Early Losses Hit Harder
Three effects compound simultaneously:
Year 1: ₹3 Cr portfolio
↓ -30% market crash
₹2.1 Cr
↓ -₹10.5L withdrawal
₹1.99 Cr ← all future returns compound on this smaller base
Effective withdrawal rate: 10.5L / 2.1 Cr = 5.0%
(vs the planned 3.5%)- Portfolio shrinks from the market loss
- Fixed withdrawal takes a larger percentage of the reduced portfolio
- All future compounding operates on the smaller base
A 30% loss in year 1 of retirement has more impact on a 30-year outcome than a 30% loss in year 20.
Indian Market Volatility vs US
| Metric | Nifty 50 | S&P 500 |
|---|---|---|
| Standard deviation (20-year) | ~24–26% | ~15–17% |
| Worst single-year loss | -52% (2008) | -37% (2008) |
| Average CPI inflation | 5.4% | 2.8% |
| Reliable equity data from | ~1990s | ~1920s |
Higher volatility means wider spreads in Monte Carlo outcomes. Higher inflation means withdrawals grow faster in nominal terms, amplifying the impact of early losses.
Monte Carlo Success Rates by Withdrawal Rate
Indian parameters: 12% mean equity return, 25% standard deviation, 5.4% inflation, 60/40 equity/debt allocation.
| Withdrawal Rate | 30-Year Success | 40-Year Success |
|---|---|---|
| 4.0% | ~78–82% | ~60–65% |
| 3.5% | ~88–92% | ~75–80% |
| 3.0% | ~95–97% | ~88–92% |
| 2.5% | ~98–99% | ~95–97% |
These shift with asset allocation, return assumptions, and inflation assumptions.
What Monte Carlo Doesn’t Capture
| Limitation | Why It Matters |
|---|---|
| Returns aren’t truly random | Markets exhibit momentum and mean reversion |
| Correlations change during crises | Equity and debt can fall together (2008, 2020) |
| Inflation isn’t independent of returns | High-inflation periods correlate with specific market regimes |
| Model depends on input assumptions | Different mean/std dev assumptions produce different success rates |
| Historical data is limited for India | ~30 years of reliable equity data vs ~100 for the US |
How WealthSim Handles This
WealthSim uses deterministic cash flow simulation — it projects your actual income, expenses, and life events year by year against a return assumption. This shows the trajectory of your net worth over time and makes it visible where and when cash flow turns negative.
The approach differs from Monte Carlo: instead of running thousands of random paths, it lets you manually adjust inputs (return rate, inflation, expense changes, career breaks) and immediately see how the projection changes. This makes it a tool for exploring specific scenarios rather than generating probability distributions.
Explore your own scenarios at app.wealthsim.in.